Deep Dive into Mortgage Data Claims Clear Racial Differences



According to an in-depth analysis of 2.4 million purchase loan applications by The Markup, a nonprofit newsroom that “oversees big technology.”

The analysis found that across the country, black applicants were 80 percent more likely to be denied a regular mortgage eligible for Fannie Mae and Freddie Mac support compared to whites with similar qualifications.

Markup analysis of data on loans for 2019 found that Hispanic home buyers were 40 percent more likely to get bounced, and Asia-Pacific Islanders were 50 percent more likely to get bounced than whites. Native American homebuyers were 70 percent more likely to be rejected than whites.

“In each case, potential borrowers of color looked almost the same on paper as white candidates, except for their race,” the study authors said.

In addition to national results, The Markup reported racial differences in 89 metropolitan areas, including New York, Los Angeles, Chicago, Houston, Atlanta, Dallas, Phoenix, Minneapolis, and Washington DC.

Markup also highlighted seven lenders with the largest variation in applicant color: DHI Mortgage Company, Lennar Mortgage, Pulte Mortgage, Freedom Mortgage Corporation, Movement Mortgage Corporation, Fairway Independent Mortgage Corporation, and Navy Federal Credit Union.

All of these lenders told The Markup that they comply with fair credit laws, and some have challenged the publisher’s analysis.

Details of The Markup’s analysis of local markets and individual lenders and the methodology used have been published separately (How We Investigated Racial Differences In Federal Mortgage Data).

Conclusions of the Issues of the Credit Sector Groups

Credit industry groups have critically assessed The Markup’s methodology, saying its analysis did not take into account the credit ratings of borrowers and did not analyze government-backed FHA, VA and USDA mortgage applications.

“A person’s credit history can help explain why seemingly comparable candidates may not always have the same lending outcome,” the American Bankers Association told The Markup. written statement… “The margin analysis not only does not take into account credit history, but also does not take into account the millions of mortgages issued by the Federal Housing Administration and other government credit programs. These programs are specifically designed to serve low- and middle-income families most at risk of mortgage denial. This glaring omission paints an incomplete picture of the mortgage market. “

The Mortgage Bankers Association provided similar review in The Markup before publication. After publishing the analysis on Wednesday, MBA denounced it as “not only deeply flawed, but clearly biased in its premise” and distorting “the problems and solutions needed to solve very serious problems that lead to unequal outcomes associated with black home ownership and wealth accumulation”.

Although the Housing Mortgage Disclosure Act (HMDA) requires lenders to provide credit rating data at the loan level to regulators, the Consumer Financial Protection Bureau is removing this information from public data, “in part due to lobbying by the mortgage industry to remove it, citing on the confidentiality of borrowers. “, – note the authors of The Markup analysis.

Authors Emmanuel Martinez and Lauren Kirchner – who both have extensive experience in data-driven investigation reporting – said they were also unable to analyze decisions made by Fannie Mae and Freddie Mac’s underwriting algorithms. Like credit ratings, lenders report these decisions in their HMDA filings, but they are not available in the public domain.

Are credit ratings and algorithms to blame?

However, Martinez and Kirchner concluded that credit ratings, along with the algorithms used by individual lenders and Fannie and Freddie, were the likely culprits for the inconsistencies they said their analysis revealed.

The classic FICO score that Fannie and Freddie demand from lenders, they said, “is widely considered harmful to people of color because it rewards traditional credit, which white Americans have more access to. It does not account for timely payments for rent, utility bills, and cell phone bills, among other things, but will downgrade people’s ratings if they are delayed and sent to debt collectors. Unlike later models, it punishes people for past medical debts, even if they have since been paid. “

Some of the raw data taken into account by Fannie and Freddie’s underwriting algorithms, including assets, employment status and debt, can also disadvantage people of color, Martinez and Kirchner noted, citing other studies.

“This is a relatively new world of automated underwriting mechanisms that deliberately cannot differentiate but are likely to do so in effect,” David Stevens, former president and chief executive officer of the Mortgage Bankers Association, told The Markup.

Stevens and others told The Markup that outsiders have little understanding of the inner workings of algorithms used not only by Fannie Mae’s Desktop Underwriter and Freddie Mac Loan Prospector systems, but similar software used by FHA, USDA, and individual banks and lenders.

Fannie Mae – which recently announced he plans to start taking timely lease payments into account when evaluating borrowers with thin credit files – he provided The Markup with a written statement saying the Desktop Underwriter analyzes loan applications “without regard to race.”

Both Fannie and Freddie “stated that their algorithms are regularly reviewed for compliance with fair credit laws both internally and at FHFA and the Housing and Urban Development Department,” The Markup reported. “HUD has emailed The Markup that it has asked the couple to make changes to the underwriting criteria as a result of these reviews, but is not disclosing the details.”

Possible new rules

In a written statement provided by The Markup, the MBA said regulators have access to a broader set of data and have not been bothered by HMDA data in the past.

“The Federal Reserve, CFPB and other regulators have made it clear that inconsistencies in denial of HMDA data are not, in themselves, determinative when it comes to assessing fair credit,” said the MBA. “Credit Fair Tests include a much broader set of information about loans and borrowers.”

V new HMDA data analysis 2020however, the CFPB concluded that black and Hispanic borrowers “continue to have fewer loans, are more likely to be denied than non-Hispanic white and Asian borrowers, and pay higher average interest rates and overall loan costs. It is clear from this data that our economic recovery from the COVID-19 pandemic will not be sustainable if it remains uneven for mortgage borrowers of color. ”

And The Markup noted that Biden’s new administration secretary for housing, Marcia Fudge, recently said Axios she believes that one of the reasons the share of black homeowners has declined is that “we never fully implemented the Fair Housing Act.”

In June, HUD reported that plans to restore the rules Introduced by the Obama administration in 2013 to address “scattered impacts,” discriminatory practices that are unintentional but nonetheless unjustified. This can create legal liability for lenders who use artificial intelligence and algorithms to evaluate borrowers.

Federal regulator Fannie and Freddie, the Federal Housing Finance Agency (FHFA), wants mortgage giants to meet new targets that support lending in minority census areas. The FHFA, which requires that at least 35 percent of purchased mortgages secured by Fannie and Freddie be taken by low and very low income borrowers, is looking for comments on the proposed targets for 2022-2024

Email Matt Carter


Source link